This application is part of a 5-year proposal for a N1DA Center that will focus on testing innovative pharrrnacotherapies for the treatment of cocaine dependence. Cocaine dependence when complicated by alcohol dependence has been associated with higher levels of cocaine use, more psychopathology, poorer retention in treatment, and a worse prognosis than either cocaine or alcohol dependence alone. This 5-year project will evaluate in a double-blind, placebo-controlled, 12-week trial the efficacy of the combination of naltrexone and disulfiram for treating cocaine dependence by reducing excessive alcohol drinking. One medication will be 100 mg/day naltrexone and the other will be 250 mg/day disulfiram. Both medications are FDA-approved in the US for alcohol dependence. The study will be a 4-cell design (52Ss per cell) where each of the two medications will be given alone, or in combination, and compared to placebo. The project proposes to randomize over a 5-year period, 208 treatment-seeking Ss with a DSM-IV diagnosis of cocaine and alcohol dependence to one of four medication/placebo groups. All medications will be given in conjunction with weekly therapy visits of psychosocial treatment. The behavioral intervention we will use is Cognitive Behavioral Coping Skills Therapy (CBT) with medication management. All Ss will receive their treatment over a 12-week period at the University of Pennsylvania Treatment Research Center (TRC)--a substance abuse outpatient treatment research program that is part of the Penn/VA Center for Studies of Addictions. Following the treatment trial, there will be two visits at 6 and 9 months post-treatment entry. Weekly patient self-reports, twice-weekly urine screens, monthly collateral reports and liver enzyme data will be obtained throughout the trial. Medication compliance will be monitored by weekly pill counts and the presence of riboflavin in the urine. When the blind is broken at the end of the study, a secondary medication compliance check will be done for Ss who received active medication by examining two in-trial blood specimens for the presence of disulfiram and/or beta-naltrexol a metabolite of naltrexone. Audiotapes of all psychosocial treatment sessions will be obtained and a subgroup will be randomly selected for independent ratings of clinician adherence. Therapeutic process will be assessed with motivation (URICA), therapeutic alliance (WAI), and acquisition of coping skills (ASRPT). In sum, we will attempt to determine whether a combined medication regimen can reduce both cocaine and alcohol use in cocaine-alcohol dependent Ss.